Noise Pollution Analysis
Agent skills to document home noise pollution and analyse recordings at the frequency level: hums, tones, low-frequency drone, cycling.
About This Project
Noise-Pollution-Analysis provides a structured approach to identifying and measuring household noise nuisances. By leveraging audio processing tools including ffmpeg, SoX, and aubio, the project enables users to perform spectral analysis on recordings to isolate specific characteristics such as hums, tones, low-frequency drones, and cycling patterns. It automates the extraction of technical metrics, including 1/3-octave band levels, harmonic families, and periodicity, to provide empirical data on environmental sound disturbances.
The project is designed for individuals seeking to quantify home noise issues through a standardized workspace and analysis pipeline. Using a Python-based script managed via uv, it transforms raw audio samples into detailed reports, including before-and-after spectrograms and background noise subtraction. This allows users to move beyond anecdotal observations and generate actionable data regarding the nature and frequency-level impact of persistent noise sources.